[Webkit-unassigned] [Bug 129803] text selection highlight is visually incorrect under heavy padding constraints

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=129803





--- Comment #11 from Mario Sanchez Prada <mario at webkit.org>  2014-03-14 06:34:12 PST ---
(In reply to comment #10)
> (From update of attachment 226101 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=226101&action=review
> 
> > LayoutTests/platform/mac/editing/selection/inline-closest-leaf-child-expected.txt:35
> > -selection start: position 2 of child 0 {#text} of child 1 {SPAN} of child 4 {DIV} of body
> > -selection end:   position 5 of child 2 {#text} of child 1 {SPAN} of child 4 {DIV} of body
> > +caret: position 0 of child 0 {#text} of child 1 {SPAN} of child 4 {DIV} of body
> 
> Looks like this is a regression. r- due to the lack of description as to why new result is correct.

You are right, sorry for not noticing it while uploading the patch.

However, I think that such a regression is not actually caused by this patch, which seems to be unveiling the issue instead. I say this because I've manually loaded and experimented with the same test in Firefox and Blink (through a recently built version of chromium which includes this patch) and in both cases the behaviour is the expected one: I can select individual characters from "ipsum" by dragging the mouse in the yellow area below.

However, if I try to do the same thing in WebKit, I get different results depending on whether this patch is applied or not (but both seem to be wrong):

* Without the patch: I'm able to select characters by dragging in the yellow area above "ipsum", but the blue selection rectangle drawn is incorrect (height is too much, invading the yellow area).

* With the patch: I can't select anything at all by dragging in the yellow area above "ipsum", which is also wrong. This does not happen in Blink with the same patch applied, though.

As mentioned before, this backport patch addresses the first issue (selection rectangle incorrectly drawn) above. Unfortunately, it also unveils the second issue, which I believe is not related to this change since that's not happening in Blink after the very same changeset has been incorporated (I can select dragging in the yellow area && the selection rectangle is properly drawn, as in FF).

So, perhaps the right thing to do here would be, instead of rebaselining those expectations (which I agree is wrong), to file a bug for the selection while dragging in the yellow area and add that test to the TestExpectations files.

What do you think?

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.